OTCStreaming Data Queries: Real-Time and Historical OTC Analytics, No Data Cap
Blog
Live prices, greeks, scenarios and time series on your OTC credit universe, with no data cap, for your users and the AI agents they work with. OTCStreaming does the computation, so your team can focus on investment strategies and risk management.
AI never stops asking
An AI agent doesn't ask one question. It asks thousands, at machine speed, often the same ones and often slightly different. If each question becomes a new pricing calculation, a few agents are enough to saturate a server. Answers slow down, and then the service stops responding for everyone, including the live prices your desk relies on.
We hate limiting users. We hate data caps. We want every user to have fast access to all their OTCStreaming data.
How it works inside
High-frequency engineering, applied to OTC analytics.
Conflation, state caching, correlation and channels are how high-frequency market-data systems keep up with exchange feeds. OTCStreaming applies them to the pricing itself: not to quotes, but to calibrated models, prices, greeks and scenarios.
Bursts flattened, stale events skipped. Events arriving during a computation fold into the next one: a burst becomes one pass on the latest state. Incoming quotes are never sampled: every distinct quote is stored.
The state, kept in memory. Each state machine keeps its latest state. Every question reads it with no new computation, and only the market updates it.
Correlated securities in one vector. Securities that move together, on the same model, are computed in one vectorised pass: prices, greeks and scenarios together.
Light and heavy updates kept apart. Each kind of update runs in its own channel, with its own processors and its own pace of conflation.
In practice, your systems and agents subscribe once and get each change as it happens. When two users ask the same question, they share one computation. Our servers work when the market moves; a new question just reads what is already computed. And because the universe of plausible credit securities is finite, that work stays bounded.
A live tensor of analytics, so you can focus on strategy and risk
Think of what OTCStreaming provides as one large tensor that is always up to date. You can slice it any way your strategy or your risk process needs.
securities × measures × scenarios × time
CDS, indices, index options, tranches, bonds, and whole curves and volatility surfaces.
Prices, spreads, volatilities, CS01, jump-to-default, carry and roll-down, on each side of the market.
Spread shocks, curve moves and defaults, computed with the price rather than after it.
Live; London Close history, following each roll as it happened; replay of today's book on any past date. History or replay can run up to the live level in one query.
Calibrating curves, skews and volatility surfaces, keeping tens of thousands of valuations and their greeks up to date, and handling rolls, history and replays is all done for you. When a model moves, every value that depends on it is updated at once, prices and risk together, so your strategies and risk reports always start from the same numbers.
Delegate the computation to OTCStreaming. Spend your time on what creates value: designing investment strategies and managing risk.
Ask once
Name the CDS, indices, options or tranches, or a whole curve or surface, and what you need: spreads, prices, volatilities, CS01, jump-to-default, carry and roll-down, scenarios.
We keep it current
OTCStreaming prices it in the background on its own models and reprices every time a model recalibrates. Prices and risk are recomputed together.
You hear when it moves
Your system is told the moment a valuation changes and receives only what changed. It doesn't ask again and doesn't reload everything.
Example: a credit desk built on OTCStreaming Data Queries
Here is how a credit trading desk could connect its systems to one OTCStreaming instance. Each system makes its own data queries, and all of them share the same computation.
Trading screens
Single-name CDS curves, index skews and option volatility surfaces, on the tenors and strikes the desk quotes. Screens move when the model moves.
Risk system
The book's positions with CS01, jump-to-default, carry and roll-down and spread scenarios, updated with the price. Only positions whose model moved are resent.
End-of-day and P&L
London Close series for every position, each date priced with the contract that was current then. Reports start from the same numbers the desk saw.
Research
Today's book priced on every day of the past year in one fast pass, for backtests and what-if studies.
For your desk
- Risk as fresh as priceGreeks and scenarios come in the same pass as the price.
- Faster reactionScreens and risk move as soon as the market does.
- Lighter integrationOne connection, small updates, less code.
- No data capsPull what your strategies and your AI need.
- ReliableAfter a disconnection or a restart, your systems resync on their own.
- AffordablePriced once per market move, shared by every user of your instance.
Available now
OTCStreaming Data Queries are an addition to the current offer, at no additional charge, including the bundle offer.
Data queries are part of the OTCStreaming API (/api/v1/data-query). Ask your OTCStreaming contact for a walkthrough, or start with the data-query section of your instance's API documentation.